Short-Form Video & Reels
7 Short-Form Video Metrics: Does It Work for B2B?

Joshua Jackai
15 min

Published 12 November 2026 · Last updated and fact-checked November 2026
Short answer: Short-form video works for B2B reach, and it can be measured there. In LinkedIn's 2024 B2B Marketing Benchmark, 55% of marketers said short-form social video produces their highest ROI, but that is an opinion survey. Pipeline effects need a holdout test, and no peer-reviewed B2B-only trial exists yet.
A head of marketing at a 60-person US software firm gets the same question from the CEO every quarter: is the video budget doing anything? The clips get views. The CRM shows no field called "came from a Reel". So the honest answer to "does short form video work for B2B" splits in two. It works for things you can count on the platform. It is unproven for things you can only count in the CRM.
This post is not a production guide. My webinar repurposing guide covers cutting clips from long recordings, and my podcast video editing guide covers the edit. Here I sort seven short-form video metrics by what each one can prove. I read every vendor page on 12 November 2026 and dropped any figure I could not find at its source.
Key takeaways
LinkedIn's 2024 benchmark, run by Ipsos with 2,001 B2B leaders, found 55% say short-form social video produces their highest ROI. It is a survey answer, not a return rate.
In the same LinkedIn report, only 16% named video as the most effective content type, so "highest ROI" and "most effective" are different answers from the same people.
LinkedIn reports 1.6x higher Lead Gen Form open rates when Video Ads run alongside. A form open is not a lead.
Lewis and Rao (2015) showed ad returns are hard to measure even across 25 large field experiments, which is the core problem for video and pipeline.
At Y Scouts, an EOS®-run executive search firm, my short-form program contributed to inbound MQLs rising 76.5% quarter over quarter. That is a correlation, not a controlled test.
I found no peer-reviewed trial of short-form video limited to B2B buyers.
Metric | What it can prove | Who measured it | Headline figure |
|---|---|---|---|
1. Views and reach | People saw the clip | LinkedIn platform data | Viewership +36% in 2024 |
2. Brand recall | Marketers believe it | LinkedIn sentiment survey | 93% agree |
3. Self-reported ROI | What marketers think pays | LinkedIn and Ipsos | 55% |
4. Lead form engagement | More form opens | LinkedIn internal data | 1.6x |
5. MQLs and demos | Movement next to video | Demand Designer, Y Scouts | +76.5% MQLs |
6. Revenue attribution | Very little on its own | Peer-reviewed field studies | Intervals over 100 points wide |
7. Holdout test | Cause, if powered | Peer-reviewed method | One control group |
1. Views and reach: best for proving people saw the clip
Views are the one short-form video number every platform counts the same way you do. LinkedIn says video viewership on its platform rose 36% in 2024 to 154 billion views. That proves attention exists on the channel. It does not prove your buyers watched, and it suits a team that needs a weekly reach number.
Source: LinkedIn, The Art and Science of Video Storytelling, 2025
Metric: video views on LinkedIn, platform-wide
Figure: +36% in 2024, to 154 billion views
Proves: the channel has video attention
Does not prove: that your target accounts watched
Reach is where short-form video is easiest to measure and easiest to oversell. A view on LinkedIn counts after a short play, and a view on Instagram Reels or YouTube Shorts counts on a different rule. So when a report adds views across platforms, it adds unlike things. I keep views per platform and never total them.
The useful cut is views from the right companies. LinkedIn page analytics shows viewers by company size and industry. If 80% of your views come from students and job seekers, the clip worked for reach and failed for B2B. That split is the first thing I check on any short-form video strategy.
Views also give you a cheap creative test. Ten clips with ten different opening lines tell you which hook holds attention, in days rather than quarters. That is real value, and it is measurable, but it is a creative signal, not a pipeline one.
"I keep views per platform and never total them," said Joshua Jackai, founder of Demand Designer.
Data point: LinkedIn's Art and Science of Video report says video viewership on LinkedIn surged 36% in 2024 to reach 154 billion views (LinkedIn platform data, not peer reviewed).

2. Brand recall: best for knowing what marketers believe
LinkedIn reports that 93% of B2B marketers agree video is better at driving brand recall than single image ads. That is a belief held by marketers, measured by asking them. It is not a recall test on buyers. It suits a team building a case for budget, not a team proving results.
Source: LinkedIn 2025 B2B Marketer Sentiment Research
Sample: not stated on the page
Metric: agreement with a statement
Figure: 93% of B2B marketers agree
Proves: marketer opinion
Recall is measurable, just not this way. A brand lift study shows an ad to one group, holds it back from another, and asks both groups later whether they remember the brand. LinkedIn, Meta and YouTube each sell versions of this for paid campaigns. Organic clips cannot get one, because the platform cannot hold anyone back from an organic post.
So the 93% tells you something about your peers, which is still worth knowing. If you present it to a CFO, call it a survey of marketers. The moment you call it proof that video drives recall, a sharp reader will ask "measured how?" and the answer is "we asked marketers".
For organic short-form video clips, the closest recall proxy I trust is branded search. If people search your company name more often in the weeks after a series runs, that is a weak but honest signal. It is still a correlation.
Data point: LinkedIn's video measurement page states that 93% of B2B marketers agree that video is better at driving brand recall than single image ads, citing its 2025 sentiment research (vendor survey, not peer reviewed).

"Observational methods often fail to accurately recover the treatment effects generated from randomized advertising experiments on Facebook," Brett Gordon of Northwestern's Kellogg School and co-authors wrote in Marketing Science in 2019.
3. Self-reported ROI: best for reading the 55% correctly
The most quoted short-form video content statistic is LinkedIn's 55%. In its 2024 benchmark, run by Ipsos with 2,001 B2B leaders, 55% said short-form social videos produce their highest ROI. It is the share of respondents who picked that format, not a 55% return. It suits a team checking whether peers rate the format.
Source: LinkedIn B2B Marketing Benchmark 2024, fieldwork by Ipsos
Sample: 2,001 B2B leaders, including 250 in North America
Metric: share naming short-form social video as highest ROI
Figure: 55%
Same report: video named most effective content by 16%
Several articles on page one of Google repeat this as "55% ROI". That is a misreading. The figure counts people, not dollars.
The same report holds a second number that rarely gets quoted. Asked which content is most effective, respondents picked video at 16%, visual content at 15%, thought leadership blog posts at 14% and case studies at 14%. The report itself says "no clear winner emerges". So marketers rate short-form video highly on ROI and only narrowly ahead on effectiveness.
The report also says roughly 2 in 5 respondents are under pressure to prove ROI in a short period and cannot measure activity between buyer stages. That is the gap this whole post is about. When the people answering the survey cannot measure stage-to-stage movement, their ROI answer is an estimate.
I also checked the 67% "short-form content is most valuable" figure that competitors credit to Demand Gen Report. Its pages blocked my reader, so I left it out.
Data point: LinkedIn's 2024 B2B benchmark reports that Ipsos recruited 2,001 B2B leaders and that 55% say short-form, social videos produce the highest ROI (vendor survey, not peer reviewed).

"The median confidence interval on return on investment is over 100 percentage points wide," Randall Lewis and Justin Rao wrote in The Quarterly Journal of Economics in 2015.
4. Lead form engagement: best for paid video on LinkedIn
LinkedIn says advertisers see 1.6x higher Lead Gen Form open rates when those forms are combined with Video Ads. This is the first metric that sits close to a lead. It counts form opens, not submissions, and it suits a team already paying for LinkedIn lead ads.
Source: LinkedIn internal data, 2025
Method: not published
Metric: Lead Gen Form open rate
Figure: 1.6x higher with Video Ads
Proves: more people opened the form
A form open is a click on "Apply" or "Download". The person still has to submit. LinkedIn's page does not say how submissions or lead quality changed, or how the comparison groups were picked. Advertisers who run video may also differ from those who do not, in budget and in audience.
Still, this is where paid short-form video earns its place for B2B. A paid clip with a native form puts the lead step one tap away, and LinkedIn records the submission against the campaign. That gives you a cost per lead you can compare with your other ads.
What I check: cost per submitted form, then the share of those leads your sales team accepts. If video forms open more but sales rejects them more, the 1.6x is not your number. Run both video and single image ads to the same audience and the same form for at least four weeks.
"A form open is not a lead," said Joshua Jackai, founder of Demand Designer.
Data point: LinkedIn's measurement page says advertisers see 1.6x higher Lead Gen Form open rates when combined with Video Ads, citing LinkedIn internal data from 2025 (vendor data, not peer reviewed).

5. MQLs and demos: best for B2B teams with a CRM and patience
This is the first-party case. At Y Scouts, an executive search firm that runs on EOS®, I ran a short-form video program in Q3 2025. Inbound MQLs rose 76.5% over Q2 and demos scheduled hit an annual high of 37. It suits a team that can wait one sales cycle and read the CRM.
Source: Demand Designer first-party, Y Scouts, Q3 2025 against Q2 2025
Inbound MQLs: 30 in Q3, +76.5% over Q2
Demos: 37 scheduled, 29 completed
Contracts won: +85.6% from Q3 to Q4
Top of funnel: YouTube views +2,721%, Instagram views +641.8%
I am careful with these numbers. Executive search runs a 30 to 90 day sales cycle, so Q3 video maps to Q4 contracts by timing and by a stated attribution model, not by a controlled test. The video contributed to the rise; I cannot say it caused it. Sales, referrals and seasonality moved in the same window.
What made the numbers readable was the CRM setup, not the clips. Every form asked how the person heard about Y Scouts, every video linked to a tagged URL, and the HubSpot report grouped MQLs by first touch. Without those three, the same program would have shown views and nothing else. The full write-up is on the Y Scouts video engine project page.
Five of the companies I have shipped work for run on EOS®, Y Scouts among them. That matters here only because EOS teams already review a weekly scorecard, so a video metric with a clear owner gets read.
Data point: On the Y Scouts program, inbound MQLs went from 17 to 30 in Q3 2025, a 76.5% rise over Q2, per the Y Scouts project page, alongside 37 demos scheduled in my client records (first-party, correlation, not peer reviewed).

"The video contributed to the rise; I cannot say it caused it," said Joshua Jackai, founder of Demand Designer.
6. Revenue attribution: best for knowing what it cannot show
Attribution reports in HubSpot or Salesforce will credit short-form video with revenue if a clip was a touch. That credit is a rule, not a measurement. Peer-reviewed field studies show even large advertisers struggle to measure ad returns. It suits a team deciding how much weight to put on attribution dashboards.
Evidence: 25 large field experiments, Lewis and Rao, 2015
Finding: median confidence interval on ROI over 100 percentage points wide
Evidence: big field experiments at Facebook, Gordon and colleagues, 2019
Finding: observational methods often fail to match the experiments
B2B-only trials: none found
Lewis and Rao studied 25 large experiments with major US retailers and brokerages. Even with millions of customers per test, the returns were hard to pin down. A B2B firm with 400 target accounts has far less data than that.
Gordon and colleagues compared randomized experiments at Facebook with the observational methods marketers use every day. The observational methods often missed. A multi-touch report is one of those observational methods.
So when someone asks "does short form video work for B2B revenue?", the honest reply is that attribution software cannot answer it alone. It can tell you which deals touched a clip. It cannot tell you which deals would have closed without one. I still keep the report, because it shows which clips buyers saw before a demo, and that helps plan the next batch.
Data point: Lewis and Rao (2015) report in The Quarterly Journal of Economics that 25 large field experiments, collectively $2.8 million in digital ad spend, reveal that measuring the returns to advertising is difficult (peer reviewed).

"To measure the effects of advertising, marketers must know how consumers would behave had they not seen the ads," Garrett Johnson of Northwestern's Kellogg School and co-authors wrote in the Journal of Marketing Research in 2017.
7. Holdout test: best for proving cause on paid video
A holdout test splits your target accounts in two, shows paid short-form video to one half and holds it back from the other, then compares pipeline. It is the only method on this list that can show cause. It suits a team with at least a few hundred accounts and one quarter to wait.
Method: randomized holdout, also called a control group
Evidence: Kohavi and colleagues, 2008; Johnson and colleagues, 2017
Needs: an account list, paid video, CRM opportunity data
Time: one sales cycle plus the test window
Limit: organic posts cannot be held back
Kohavi and colleagues describe controlled experiments as the strongest design for linking a change to user behaviour. Johnson and colleagues built "ghost ads" so the held-back group's would-have-seen moment gets logged too.
For a B2B team, the simple version is account-based. Upload half your target list as a LinkedIn matched audience for video ads. Keep the other half out of every paid video campaign. After one sales cycle, compare opportunity rates between the halves.
The weak spot is size. With small lists, chance swings swamp the effect, which is the problem Lewis and Rao measured. My calculator in the next section shows how many accounts each half needs before the result means anything. If your list is too small, run the test across two quarters, or test on lead form submissions first, since they arrive in weeks rather than months.
Data point: Kohavi and colleagues wrote in Data Mining and Knowledge Discovery that controlled experiments "embody the best scientific design for establishing a causal relationship" (peer reviewed, 2008).

"Controlled experiments embody the best scientific design for establishing a causal relationship between changes and their influence on user-observable behavior," Ron Kohavi and co-authors wrote in Data Mining and Knowledge Discovery.
Comparison matrix: what each metric can prove
This matrix sorts the seven short-form video metrics by what they prove, who reported them and whether you can run them yourself. Use it to pick which number goes in the CEO report and which stays in the marketing channel. Every figure comes from the items, with its source linked there.
Metric | Proves cause? | Source type | You can run it? | Figure |
|---|---|---|---|---|
Views and reach | No | Platform data | Yes, native analytics | +36% LinkedIn views in 2024 |
Brand recall survey | No | Vendor survey | No | 93% of marketers agree |
Self-reported ROI | No | Vendor survey (Ipsos) | No | 55% of 2,001 leaders |
Lead form opens | No | Vendor internal data | Yes, paid only | 1.6x |
MQLs and demos | No, correlation | First-party CRM | Yes, with tagged links | +76.5% MQLs |
Revenue attribution | No | Software rule | Yes | Intervals over 100 points wide |
Holdout test | Yes, if powered | Peer-reviewed method | Yes, paid only | One control group |
Holdout calculator: how many accounts you need
The calculator in the interactive version of this page takes your current opportunity rate and the lift you hope video adds, then returns how many accounts each half of a holdout needs. It uses the standard two-group sample size formula at 95% confidence and 80% power. Default inputs are placeholders, not benchmarks.
If the answer is bigger than your target account list, a holdout cannot settle the question for you yet. Report views and lead form numbers honestly instead, and keep the test for when the list grows.
Frequently asked questions
These questions come from the top-ranking Google pages for short-form video and B2B searches in the US, and from AlsoAsked exports for the same terms. Each answer is short enough to quote and uses only the figures already sourced in the seven items, so nothing new is introduced here.
Does short-form video work for B2B?
Short-form video works for B2B reach and creative testing, and both can be measured on the platform. In LinkedIn's 2024 benchmark, 55% of B2B leaders named it their highest-ROI format, but that is a survey answer. Pipeline effects need a holdout test, and I found no peer-reviewed B2B-only trial.
How long is a short-form video?
Most platforms treat short-form video as clips under about 60 seconds, shot vertically for phones. YouTube Shorts allows up to three minutes, so the line varies by platform. For B2B, length matters less than whether the opening seconds name a problem your buyer has.
Which platform is most effective for B2B video?
LinkedIn is the platform that reports B2B video data most directly, including 1.6x higher Lead Gen Form open rates with Video Ads. YouTube suits search-led clips. Pick the platform where your target accounts already spend time, then check viewer company data before scaling spend.
What are the downsides of short-form video for B2B?
The main downside is measurement. Views are easy to count and pipeline is hard to prove, so teams often report reach as if it were revenue. Roughly 2 in 5 B2B leaders in LinkedIn's 2024 benchmark said they cannot measure activity between buyer stages.
How I set up short-form video for B2B clients
When a client asks me whether short-form video will work, I start with the CRM, not the camera. I add a "how did you hear about us" field, tag every clip link and build one report grouped by first touch. Then I plan clips from material the team already has, such as webinars and sales calls.
I make short-form video and B2B websites at Demand Designer, so weigh my view with that in mind. The 1-to-12 Engine is my free plan for turning one recording into twelve clips, and my guide to measuring social media ROI covers the reporting side.
EOS® is a registered trademark of EOS Worldwide. Demand Designer is not affiliated with or endorsed by EOS Worldwide.
Sources
Peer-reviewed research
Gordon, B. R., Zettelmeyer, F., Bhargava, N., and Chapsky, D. (2019). A comparison of approaches to advertising measurement: Evidence from big field experiments at Facebook. Marketing Science.
Johnson, G. A., Lewis, R. A., and Nubbemeyer, E. I. (2017). Ghost ads: Improving the economics of measuring online ad effectiveness. Journal of Marketing Research.
Kohavi, R., Longbotham, R., Sommerfield, D., and Henne, R. M. (2008). Controlled experiments on the web: Survey and practical guide. Data Mining and Knowledge Discovery.
Lewis, R. A., and Rao, J. M. (2015). The unfavorable economics of measuring the returns to advertising. The Quarterly Journal of Economics.
Industry, vendor and first-party sources (not peer reviewed)
LinkedIn and Ipsos. B2B Marketing Benchmark 2024 (copy hosted by PPC Land). Read 12 November 2026.
LinkedIn. The Art and Science of Video Storytelling, 2025 (copy hosted by PPC Land). Read 12 November 2026.
LinkedIn. Measure video ads. Read 12 November 2026.
Demand Designer. Y Scouts video engine. First-party client result.
Free workbook
The 1-to-12 Engine
Turn one recording into twelve short videos that are still being found months later.
Five steps, five fill-in prompts, a workbook for each. Free.







